日程：2019年1月9日(水)-11日(金)

会場：	ルスツリゾート（北海道蛇田郡留寿都村字泉川13）
	http://www.rusutsu.co.jp/
会場地図:	ノースウイング　コンベンションホール　18番ホール

スケジュール:

1月9日（水）-11日（金）
認知発達と発達障害：予測符号化の視点から"Cognitive development and its disorders : From the viewpoint of predictive coding"
9日	スペシャルセッション
18:10 - 19:00	Caroline Catmur (King’s College London)
19:10 - 20:00	橋本龍一郎 (首都大学,昭和大学)
20:10 - 21:00	James Kilner (University College London)
21:00-23:00	ポスターセッション
10日	トピックセッション
15:30 - 16:20	Jean-Jacques Slotine (Massachusetts Institute of Technology)
16:30 - 17:20	Rebecca Lawson（University of Cambridge)
17:30 - 18:20	山下祐一（国立精神・神経医療研究センター)
20:00-23:00	ポスターセッション
11日	トピックセッション
8:50-9:00	ポスター発表賞の紹介
9:00 - 9:50	大隅典子 (東北大学）
10:00 - 10:50	高橋哲也 (福井大学)
11:00 - 11:50	竹内倫徳（Aarhus University）

Abstracts and References：

Caroline Catmur
King’s College London

The impact of alexithymia and interoceptive impairments on the emotional symptoms of autism.

Alexithymia, a difficulty in identifying and describing one’s emotional states, is a sub-clinical condition with a high prevalence in a number of psychiatric conditions including autism, eating disorders, and schizophrenia. In this talk I will outline a series of studies in which we have identified the relative contributions of alexithymia and autism to various aspects of social cognition including emotion perception, theory of mind, and judgements of morality. I will then discuss how alexithymia may relate to impairments in interoception, the ability to perceive one’s internal states, and how interoceptive impairment and alexithymia relate to emotional decision-making in autism.

References :
1.Oakley, B.F.M., Brewer, R., Bird, G. & Catmur, C. (2016). ‘Theory of Mind’ is not Theory of Emotion: A cautionary note on the Reading the Mind in the Eyes Test. Journal of Abnormal Psychology, 125(6): 818-823. DOI
2.Brewer, R., Marsh, A. A., Catmur, C., Cardinale, E. M., Stoycos, S., Cook, R., & Bird, G. (2015). The impact of Autism Spectrum Disorder and alexithymia on judgments of moral acceptability. Journal of Abnormal Psychology, 124(3): 589-595. DOI
3.Shah, P., Catmur, C. & Bird, G. (2017). From heart to mind: Linking interoception, emotion, and theory of mind. Cortex, 93: 220-223. DOI
4.Murphy, J., Brewer, R., Hobson, H., Catmur, C. & Bird, G. (2018). Is alexithymia characterised by impaired interoception? Further evidence, the importance of control variables, and the problems with the Heartbeat Counting Task. Biological Psychology, 136: 189-197. DOI
5.Shah, P., Hall, R., Catmur, C. & Bird, G. (2016). Alexithymia, not autism, is associated with impaired interoception. Cortex, 81: 215-220. DOI
6.Shah, P., Catmur, C. & Bird, G. (2016). Emotional decision-making in Autism Spectrum Disorder: The roles of interoception and alexithymia. Molecular Autism, 7: 43. DOI

橋本龍一郎
Ryuichiro Hashimoto
首都大学・昭和大学
Tokyo Metropolitan University, Showa University

安静時脳機能ネットワークに基づく発達障害研究の次元的アプローチの展開
Dimensional approach in developmental disorder research based on resting-state functional brain network

Resting-state brain activity in psychiatric disorders has garnered interest from basic neuroscientists as well as clinicians possibly due to its potential to be linked with explicit models for spontaneous computational processes such as predictive coding. Among others, autism spectrum disorder (ASD), a major developmental disorder, is one the first psychiatric disorders in which altered default-mode activity was identified (Kennedy et al., 2006). This discovery led to a spur of resting-state fMRI studies of ASD with a particular focus on altered functional connectivity (FC) over various developmental stages. However, the exact characterization of the abnormalities has been unclear probably due to difficulties inherent to developmental disorders including altered courses of brain development and population heterogeneity. In this talk, I review recent progress in our understanding of characteristics of resting-state fMRI in autism over multiple developmental stages, with a particular focus on findings from high-functioning ASD in the adulthood (e.g., Itahashi et al., 2014). Here, I will illustrate a resting-state FC study of graph-theoretical analysis for the adult ASD, which will be compared with results obtained by similar approaches for pediatric populations. I will then describe our recent efforts to develop resting-state FC-based biomarker for ASD (Yahata et al., 2016). Although the project had been initially motivated by the need for high accuracy biomarkers for distinguishing between ASD and the typically developed population, the application of the classifier to other psychiatric populations revealed unexpected findings of overlaps between ASD and schizophrenia. This observation leads to the last topic of heterogeneity and cross-disease overlap within the neurodevelopmental disorder spectrum. There, I will review evidence for overlaps in behavioral and neuroendophenotypes including predictive coding between two major developmental diseases of ASD and ADHD (e.g., Aoki et al., 2018), which will be ensued by a discussion about the possibility for neuroimaging-informed stratification of the developmental disorders to better understand their neuronal substrates and to develop the interventions.

References :
1. Kennedy, D.P., E. Redcay, and E. Courchesne: Failing to deactivate: resting functional abnormalities in autism. Proc Natl Acad Sci U S A, 103(21): 8275-80 (2006)
2. Itahashi, T., Yamada, T., Watanabe, H., Nakamura, M., Jimbo, D., Shioda, S., Toriizuka, K., Kato, N., & Hashimoto, R.: Altered network topologies and hub organization in adults with autism: A resting-state fMRI study. PLoS ONE, 9(4) e94115: 1-15 (2014)
3. Yahata, N., Morimoto, J., Hashimoto, R., Lisi, G., Shibata, K., Kawakubo, Y., Kuwabara, H., Kuroda, M., Yamada, T., Megumi, F., Imamizu, H., Nanez, J., Takahashi, H., Okamoto, Y., Kasai, K., Kato, N., Sasaki, Y., Watanabe, T., & Kawato, M.: A small number of connections predicts adult autism spectrum disorder. Nature Communications 7(11254): 1-12 (2016)
4. Aoki Y, Yoncheva YN, Chen B, Nath T, Sharp D, Lazar M, Velasco P, Milham MP, Di Martino A. Association of White Matter Structure With Autism Spectrum Disorder and Attention-Deficit/Hyperactivity Disorder. JAMA Psychiatry. 74(11): 1120-1128 (2017)

James Kilner
University College London

The possible role of sensory uncertainty in movement and movement disorders

Every movement we make stimulates peripheral sensory receptors that provide sensory feedback of the motor act. It is thought that when we move we predict the sensory consequences of that movement (through forward models) and compare this prediction to the actual sensory input. Any difference between the predicted and actual sensory input will result in a prediction error. In order to determine the relevance of any prediction errors, the model requires estimates of both the uncertainty in the motor prediction and the uncertainty of the actual sensory input. The importance of the estimate of uncertainty at both of these levels is highlighted in a recent theoretical account of motor control and movement initiation: active inference. In active inference sensory attenuation prior to and during active movement is an essential step in actually being able to move. Of particular interest here is that within the active inference framework a failure to move can be modelled by a failure to sufficiently attenuate precision on the somatosensory expectations. Indeed, it has been proposed that some of the hypokinetic symptoms of Parkinson’s disease, specifically akinesia and bradykinesia, can be recast as a result of a pathology in reducing the precision of the somatosensory expectations.

Jean-Jacques Slotine
Massachusetts Institute of Technology

Collective computation and learning in nonlinear networks

The human brain largely outperforms robotic algorithms in most tasks, using computational elements 7 orders of magnitude slower than their artificial counterparts. Similarly, current large scale machine learning algorithms require millions of examples and close proximity to power plants, compared to the brain's few examples and 20W consumption. We show that nonlinear systems tools, such as contraction analysis and virtual dynamical systems, yield simple but highly non-intuitive insights about collective computation in networks, and in particular the role of sparsity, and that they also suggest systematic mechanisms to build progressively more refined networks and novel algorithms through stable accumulation of functional building blocks and motifs. Imposing contraction and physics-based constraints may yield orders of magnitude improvements in computational efficiency in adaptive robotics and general machine learning. We discuss specifically contraction analysis of networks of natural gradient learners, asynchronous distributed adaptation, multiple time-scale primal-dual optimization, and linearization-free simultaneous localization and mapping.

Rebecca Lawson
University of Cambridge

A more precise look at "context" in autism spectrum disorder

Autism is a neurodevelopmental condition with a complex genetic basis that affects how people interact with the social and the non-social environment. Prominent psychological theories of autism have suggested that the core cognitive difficulty concerns "context insensitivity", a tendency towards small details at the expense of global processing. Predictive coding theories offer a computational account of how neural processing is contextualised, via precision or neural gain. In this talk I will first outline a predictive processing account of autism spectrum disorder. I will then present the findings from a series of recent studies examining differences in how autistic people process and learn about different kinds of context, leading to (in)flexible decisions in a changing world.

山下祐一
Yuichi Yamashita
国立研究開発法人　国立精神・神経医療研究センター
National Center of Neurology and Psychiatry

階層的予測プロセスの失調としての精神障害・発達障害：神経ロボティクス的アプローチ
Psychiatric and developmental disorders as failures in hierarchical predictive process: neurorobotics approach

The idea that adaptive behavior and cognition of human are enabled by hierarchical predictive process under generative models (i.e. predictive coding) has become a popular perspective to explain computational principle of the brain. However, while hierarchical predictive process provides significant advantages for adaptive behavior in social environments, their failure to properly develop or maintain precisely aligned signaling in neural systems has been postulated to result in psychiatric or developmental disorder symptoms.
In this talk, I will introduce our series of experiments in which behavioral control mechanisms with hierarchical predictive process were implemented by the physical actions of a humanoid robot driven by a hierarchical recurrent neural network. These experimental results demonstrated that failures in hierarchical predictive process including functional disconnection between levels of the hierarchical network and altered estimation of sensory precision lead to aberrant neural activities and abnormal behavior that can explain psychiatric and developmental disorders symptoms. We suggest that our methodology of a neural network model-driven robot could provide us new insights for examining the hypothesis of pathological mechanisms in psychiatric and developmental disorders.

References :
1. Yamashita Y, Tani J (2012) Spontaneous Prediction Error Generation in Schizophrenia. PLoS ONE 7(5): e37843.
2. 山下祐一、松岡洋夫、谷淳 (2013) 計算論的精神医学の可能性：適応行動の代償としての統合失調症,　精神医学 55: pp885-895.
3. Idei H, Murata S, Yamashita Y, Tani J and Ogata T (2018) A Neurorobotics Simulation of Autistic Behavior Induced by Unusual Sensory Precision, Computational Psychiatry (in press)

大隅典子
Noriko Osumi
東北大学
Tohoku University

Paternal aging induces atypical behaviors in offspring: a model for neurodevelopmental disorders and possible underlying mechanisms

Epidemiological studies consistently suggest strong association between paternal aging and incidence of neurodevelopmental disorders such as autism and schizophrenia in offspring. We have previously shown that a single gene mutation can cause different behavior phenotypes due to paternal aging (Yoshizaki et al., 2017). Therefore, we are now challenging to reveal underlying molecular and cellular mechanisms. We found that F1 offspring of aged fathers showed lower body weight at the early postnatal stage, as well as impairment in pup’s vocal communication, sensorimotor gating and spatial learning. We investigated in more detail ultrasonic vocalization (USV) induced by maternal separation of pups because the data obtained are considered to be suitable to analyze developmental trajectory during early postnatal stages. The overall syllables of pups derived from aged fathers showed reduction in the number and duration, but different developmental pattern in the interval. The pups of aged fathers showed increased the percentage of the simple syllable such as “downward”, conversely decreased the percentage of complex syllables such as “wave”, “chevron”, “more jump” and “more jump + harmonics”. Furthermore, entropy analyses revealed that pups of aged fathers emitted significantly narrower spectrum of syllable types from postnatal day 3 to 12. Interestingly, principal component analysis found that paternal aging increased the individuals with atypical syllable developmental patterns during early postnatal days. Regarding underlying mechanisms, we found reduction of thickness in the neocortex, especially in the deep layer, at postnatal day 6, when vocalization defects were observed. We profiled gene expression of developing brains at embryonic day 11.5 (E11.5) and E14.5 by RNA-seq. Gene Set Enrichment Analysis (GSEA) showed in brains of aged-father derived offspring at E14.5, but not in E11.5, enrichment of “Late-fetal genes” and conversely, that of “Early-fetal genes” in those derived from young father. Interestingly, “NRSF/REST motif genes” and “SFARI genes (autism-related)” were found to be highly enriched in the brains derived from aged father. A possible scenario is that paternal aging may induce in the offspring’s brain precocious neurogenesis via dis-regulation of gene expression by NRSF/REST, a pivotal transcription factor for neurogenesis. Our parallel analyses on sperm DNA methylation at the whole genome level also suggest involvement of hypo-methylation in aged sperm and NRSF/REST as a common motif in hypo-methylated regions. Our paternal aging model can be a unique tool to understand the epigenetic mechanism uderlying neurodevelopmental disorders.

References :
1. Kimura R, Yoshizaki K, Osumi N.: Risk of Neurodevelopmental Disease by Paternal Aging: A Possible Influence of Epigenetic Alteration in Sperm. Adv Exp Med Biol. 2018;1012:75-81. doi: 10.1007/978-981-10-5526-3_8.
2. Yoshizaki K, Furuse T, Kimura R, Tucci V, Kaneda H, Wakana S, Osumi N.: Paternal Aging Affects Behavior in Pax6 Mutant Mice: A Gene/Environment Interaction in Understanding Neurodevelopmental Disorders. PLoS One. 2016 Nov 17;11(11):e0166665. doi: 10.1371/journal.pone.0166665.

高橋哲也
Tetsuya Takahashi
福井大学
University of Fukui

如何にして複雑な脳内ネットワークを捉えるか
How to characterize complex brain networks

The human brain is a complex system that is characterized by its astonishing complex behavior. This complex behavior arises from an elaborate neural network system that operates over a wide range of temporal and spatial scales. Theoretical concept from mathematics and physics have been widely applied to the study of both healthy and pathological brain conditions. A new discipline called “complexity science” has been therefore emerging. Practically, the brain complexity facilitates learning and optimal environmental adaptation to the changing demands of a dynamic environment. This complexity also conveys important information about neural system dynamics and their alterations. Electroencephalogram and magnetoencephalogram (E/MEG) can directly measure brain electric and magnetic fields of the cortex with excellent temporal resolution, thereby yielding insight into temporal dynamics within physiologically relevant frequency ranges. Therefore, it is well suited to measure the complexity and neural network of the brain. In this workshop, empirical studies of the complexity and network analyses applied to E/MEG signals in respective of development, aging and their alterations will be presented.

References :
1. Hasegawa C, Takahashi T, et al. (2018) Developmental trajectory of infant brain signal variability: a longitudinal pilot study. Front Neurosci.
2. Takahashi T, et al. (2017) Band-specific atypical functional connectivity pattern in childhood autism spectrum disorder. Clin Neurophysiol.
3. Takahashi T, et al. (2016) Enhanced brain signal variability in children with autism spectrum disorder during early childhood. Hum Brain Mapp.
4. Takahashi T, (2013) Complexity of spontaneous brain activity in mental disorders. Prog Neuropsychopharmacol Biol Psychiatry.

竹内倫徳
Tomonori Takeuchi
オーフス大学
Aarhus University

選択的な記憶保持と知識の更新
Selective memory retention and updating schematic knowledge

Knowledge plays a central role in human life. Indeed, we are who we are largely because of what we learn and what we remember. Our knowledge structure (‘schema’) consists of our past experiences and facts stored in our long-term memory. We use our schemas to organize current knowledge and provide a framework for future understanding. A key but poorly understood issue is how the memories of everyday events initially stored in the hippocampus are ‘selected’ and then ‘assimilated’ into a relevant schema in the neocortex.
Selective retention can be triggered by novelty-induced dopamine release in the hippocampus. We made a ground-breaking finding (Takeuchi et al., Nature, 2016): projections from the noradrenergic locus coeruleus to the hippocampus can drive the novelty-induced memory enhancement via a non-canonical release of dopamine. This study also raises a possibility that the impact of distinct novel experiences that bear only minimal relationship to past experiences (‘distinct novelty’) may differ from novel experiences that share some commonality with past ones (‘common novelty’) (Yamasaki and Takeuchi, Neural Plasticity, 2017; Duszkiewicz et al., Trends Neurosci, In press). We now propose that memory of events accompanied by novelty can be selectively retained through two distinct dopaminergic mechanisms, depending on the nature of the novel experience itself.
Selected new memories can assimilate into the neocortical schema very rapidly if the relevant schema is already learned (Tse et al., Science, 2007). The functional mapping with immediately-early gene expression indicated that medial neocortical structures (the prelimbic, the anterior cingulate and anterior region of the retrosplenial cortices)-hippocampal connectivity was strongly associated with successful assimilation of new information into the relevant schema (Tse, Takeuchi et al., Science, 2011; Takeuchi and Tamura, unpublished). Pharmacological interventions established that there was parallel memory encoding in the neocortex, including the prelimbic cortex, through NMDA (N-methyl-D-aspartate)-type glutamate receptor-dependent plasticity mechanisms during the hippocampal-dependent learning of new information against the backdrop of a schema (Tse, Takeuchi et al., Science, 2011).
Understanding the neural mechanisms of selective retention and assimilation of selected new memories into relevant schema may bring us towards a more effective and consciously aimed behavioural schema therapy as well as provide suggestions for better teaching and learning strategies.